The AM Forum

7/8" Heliax has much better attenuation numbers than even LMR-600.  Why?

Is it just the spacing? Looking at a photo, I'm not sure the distance between the center conductor and the shield is greater because the center conductor is a tube which takes a lot of space.

Is it the insulation? Both have very low density foam.

Is it just more surface area of the conductors?

the gas inside the foam is my guess looks the center conductor looks hollow I don't care for coax that has a copper coated center conductor if you buy coax that not solid copper your going have loss there!

Deleted

The lowest loss coax has air dielectric. The loss factor for foam dielectric coax is defined by how much air is part of the mix or how many air pockets are in the foam. The lowest loss stuff is more air than foam.

That and the center conductor is quarter inch thin wall tubing.

I have heliax that has a 3/16 alum center conductor that is copper plated/coated...foam dielectric and copper shield....I have joined pcs together by soldering a short pc of  3/16 copper tube over the center conductor then put in a pc of foam over that and soldering a standard 5/8 copper pipe connector to connect the outer shields....  The analyzer doesn't show any impedance bump..

What makes Heliax good?

At least 3 things:

1..well engineered connector series
2..it is tough outdoors
3..a high velocity factor

73DG

I used to think that the type of dielectric had almost everything to do with the loss in coax. Not so.

After all, foam dielectric RG-8 has less loss that solid dielectric RG-8. However, the reason for that is largely because the center conductor has to be made smaller when we switch from foam to solid dielectric (for the same impedance), and the increases in loss are because of that decrease in the diameter in the center conductor.

Here's what WB2WIK once said on QRZ:

"Up through a few hundred MHz, the "dielectric loss" in coaxial cables is nearly "nothing" for all popular dielectrics, and is very dwarfed by the conducted loss which comes mostly from the skinny center conductor.

"The only reason air or 'mostly air' dielectric cables (including foamed PE, etc) have 'less loss' than solid dielectric cables has nothing to do with the dielectric and everything to do with the center conductor diameter.

"For a typical .405" cable like RG213/U (sized), solid PE or PTFE dielectric cable (DC for both is 2.1 ~ 2.3) will require a #13AWG center conductor to hit 50 Ohms characteristic impedance. Changing to cellular poly ("foam"), which has a lower DC, requires a #11 AWG center conductor -- bigger, fatter, lower loss. Changing to "air," to maintain 50 Ohms you can go to a #9AWG center conductor, which is bigger, fatter, and lower loss still.

"It's all in the conductor size used, not actually the dielectric itself.

"The dielectric material (and its dielectric constant) drives what gauge the center conductor needs to be.

"'Dielectric loss' using popular cable dielectrics is almost immeasurably low for all of them until you get up into the SHF range."


Furthermore, the solid inner and outer conductors in Heliax and hardline further reduce losses compared to stranded center conductor and woven shield braid.

Mike,

I appreciate the info about common coax dielectric not having much effect up until the SHF range.

I didn't realize the inner conductor on coax was that critical in size.  

I will defer to your info and have deleted my post about dielectric being the main loss factor.  

TNX, OM.

T

Good stuff. Remember, dielectric losses increase linearly with frequency and resistive losses increase as the square root of frequency.

This discussion makes me ask a question..Why was 50 ohms decided upon as a standard for transmission lines? At HF, most of the losses in coax seem to be resistive in the center conductor. One would think a 40 ohm line with a larger center conductor would be less lossy than the same diameter 50 ohm line.

One example is the RG8-X series of coax, the same diameter as RG-59 but able to handle a lot more power at HF.

Then begs the question, why was 75 ohms decided on as a defacto standard for TV VHF/UHF use, but not for RF transmission? The only major use of 75 ohms for RF transmission that I know of was the RCA FM transmitters of the 1950s, which used their own particular series of rigid line. That might have been an RCA sales scam, one *had* to use RCA transmission line with their antennas and transmitters.

Is 50 ohm line an intersection between resistive and dielectric losses, or not? If it is, why is 50 ohms used for air dielectric lines as well as plastic dielectric lines?

Ideas?

Good questions, Bill.

I would like to add one too:

Years ago I ran three hardlines to each tower - simply cuz I had the extra feedline and wanted to avoid switching.  

When running 6M I asked if I put TWO feedlines in parallel and then matched them to 25 ohms, would this reduce the loss to 1/2?  I was told that the loss would be the same but was not offered a reason why. I assumed it was because of cummulative dielectric loss, but now I'm not sure.

IE, what will happen to overall loss when two coax cables are put in parallel and then matched properly - will it change?



Steve:

"Resistive losses increase as the square root of frequency"    This is because of skin effect, no?


T

IIRC correctly 50 Ohms is a compromise between power handling and loss. The best power handling occurs at around 25-30 Ohms. The least loss is around 75 Ohms (thus small signal lines like video use this). The 25 Ohm line was too expensive or unwieldy to 50 Ohms was selected as a compromise.

25 ohm cable will also have a low breakdown voltage.
I think video is 75 ohms for the lower C per foot if I remember.

You’re right Steve.  The only book I have that touched on these optimum numbers for coaxial cable impedance is ironically a book on i.c. design titled “The Design of CMOS Radio-Frequency Integrated Circuits” by Thomas H. Lee.  Pages 141 to 143 cover this topic and go through the equations to find the optimum air-dielectric coax impedances for best power-handling and lowest loss/optimum Q.

For power handling, the number came out to 30 Ohms.  For lowest loss, the number came out to 77 Ohms.  The TV engineers rounded this number off to 75 Ohms.

Again rounding off the geometric mean of these two values (48 Ohms) gives us the 50 Ohms compromise.

You can thank Steve Katz, WB2WIK for that one.  :)

As for why 50 ohms is a standard, the last story I heard went something like this: During WWII, the developers of radar technology went to their local hardware store, bought some handy sizes of pipe for the inner and outer conductors, and it just happened to be ~50 ohms.

92Ω coax was used in AM car radio whip antennas because of the large impedance mismatch; the most important consideration was low capacitance per foot to minimize attenuation.

There's an interesting thread with lots of other explanations as to the reasons for the different coax impedances at http://forums.qrz.com/archive/index.php/t-244930.html, "Why 50Ω, Just WHO made this decision?".

Why would the lowest loss not also equate to maximum power handling capability? After all, lowest loss = lowest heating of the coax.

75 ohms would probably have lower capacitance per foot, but that would really only make a difference if there were an impedance mismatch.

Power handling is determined by conductor size. You can use larger conductors at the lower impedance.

Early TV antennas used folded dipoles which are 300 ohm.  So when coax was used on the antenna, a simple 4/1 balun resulted in 75 ohm coax.  That's why.

I think there were some other factors as to why 300 ohm antennas were used for TV reception.  IIRC I think 300 ohms made a better match to the signals natural impedance in the air or some such nonsense.  If signals even have something called natural impedance.  I seem to remember reading about that or maybe it was something I dreamed about.  In fact, I think that impedance is actually 277 ohms.

I use to dream about antennas all the time, as I installed them for about 45 years.  I also use to have many nightmares about antenna work, but I won't go into that now.

Fred

I think there's something we're missing here.

As we increase the OD of the center conductor--for a given shield ID--the impedance lowers. That's a given.

And larger center conductors have less loss at a given current. Also a given.

However, as we increase the size of the center conductor, the dielectric gets thinner.  And when that happens, we lower the voltage rating of the coax. (25Ω coax has almost no dielectric.)

But for CATV applications, and at amateur power levels, is the voltage breakdown rating a factor? I don't think so.

So why should 75Ω coax, with a smaller center conductor than 50Ω coax, have less losses than 50Ω coax, when the I²R losses of the 75Ω cable's center conductor are higher?

I haven't done the math, but perhaps the reason is the same reason that 450Ω ladder line has lower losses: for a given power level, the current is lower as we increase the impedance.

From Lampen's article in Radio World some years ago:

DIFFICULT RATIO

  One other question that often comes up: Why 50 ohms and 75
  ohms, or any other impedance, for that matter?

  These impedances were not chosen by accident. It was known in
  the 1920s that cables of different impedance worked better for
  one application that the other. For instance, it was
  determined, through experimentation, that the best power rating
  was around 30 ohms.

  Because the impedance of any coax cable is the ratio of the
  sizes of the center conductor and the distance to the braid
  (and the quality, or "dielectric constant" of the plastic in
  between), you might wonder why we don?t have 30-ohm coaxes.

  To be sure, there are customers out there who would buy as much
  30-ohm coax as we could make! The problem is that 30 ohms
  represents a ratio very difficult to make ? so much so that it
  is quite likely that most of what you would make, you would
  throw away. Only a small percentage would be usable.

  The people who would die for 30-ohm coax are the really
  high-power people. Those are customers such as nuclear
  physicists (those with some kind of atom smasher) or medical
  scanners, such as X-ray, CAT scanners or NMRI machines. They
  all use high power and would love to have 30-ohm coax. It would
  deliver their voltage with even less loss (and higher
  efficiency).

  Those customers who wanted low-signal attenuation found that
  the ideal impedance was 77 ohms. But this was an odd number in
  terms of wire sizes. If you "fudged" just a bit to 75 ohms,
  then standard wire sizes and dimensions could be used. This was
  why all those low-power, low-voltage signal-carrying cables
  (baseband video, CATV/broadband, antenna lead-in) were all
  75-ohm.

  And then there are customers who want to deliver high voltage;
  the ideal impedance for them is around 60 ohms. This is an
  eminently "makeable" cable, but it never really got started,
  mainly because it was soon realized that most high-voltage
  customers were often high-power customers too. Therefore, there
  really needed to be a compromise between voltage and current;
  that compromise was 50 ohms.

  We'll have more on the strange story of 50-ohm coax, with a
  trip into the world of wireless microphones, in the next
  installment. Let me know if you have any questions or comments.

  Steve Lampen is a senior audio video specialist for Belden Wire
  & Cable Co. in San Francisco. His book, "Wire, Cable, and Fiber
  Optics for Video and Audio Engineers," is published by
  McGraw-Hill. Contact

Some more links:

http://radioworld.com/article/bell-labs-and-coaxial-cable/17425

http://radioworld.com/article/is-there-an-ideal-impedance/17424

http://www.wired.com/thisdayintech/2009/12/1208coaxial-cable-patent/

The power rating of coaxial line is related to the amount of r-f current that it can carry without exceeding the permissible temperature rise along that transmission line including the effects of load SWR, and the environment (local ambient air temperature, solar heating, and the thermal characteristics of the insulating media within the line, itself).

For example, 75-ohm coax requires less r-f current, and hence has lower loss to convey the same amount of power for a given length and frequency, but is not capable of carrying the same maximum power as its 50-ohm version.

Andrew LDF4-75A, 1/2" OD, 75-ohm Heliax is rated for 0.273 dB loss per 100 feet and 3.8 kW maximum average power at 20 MHz into a load SWR of 1:1.

Andrew LDF4-50A, 1/2" OD, 50-ohm Heliax is rated for 0.3 dB loss per 100 feet and 7.75 kW maximum average power for the same conditions.
//

Thanks, Steve.

One of those contained this:
"In fact, the first transmission-line coaxial cables were made by taking small copper pipe and putting it inside large copper pipe. If you do this with standard sizes of copper pipe, you will get impedances like 51.5 ohms or 52 ohms."

Interesting. :-)

_______________

Just to note that early VHF and some high-power MW AM broadcast stations used 51.5 ohm rigid coaxial line, which was a standard of the RMA (Radio Manufacturers Association) in those early days.  Rumor was that this impedance was the result of the standard dimensions of pipes found in hardware stores, also, but who knows?

Later the industry moved to a 50-ohm standard whose mechanical dimensions were defined by the EIA (Electronics Industry Association).  This was valuable, as it removed the variability in the wall thickness of the inner conductors of rigid coaxial lines, the design/hole pattern of the flanges for their outer conductors, and the cutout for the anchor insulator used at each flange allowing for reduced impedance mismatches where line sections joined.

Previously all of this had caused problems with the mechanical compatibility and system performance possible when a given installation used components supplied by various manufacturers.
//

The earliest coaxial lines made by Western Electric were 66-2/3 Ohm impedance.

The center conductor was supported by porcelain spheres, which were located about a foot apart, and held in place by a small crimp on both sides of the sphere.

The 90 degree elbows and terminations were soldered on, and had little glass inspection ports to make sure it was OK.

Nitrogen or dry air ruled as well.

Typical Bell Laboratories overkill, in the pre-standard days.

I pulled a lot of it when redoing an AM directional array a few years back.

73DG

Amazing. What is the reason for this, the thermal conductivity of the dielectric? I wonder what the difference in power handling capability is for gas insulated lines of similar dimensions?

Below from Andrew Catalog 36 are those ratings for standard conditions:

   * Andrew HJ4-50, 1/2" OD, gas-insulated 50 Ω Heliax is rated for 0.368 dB loss per 100 feet and 5.36 kW maximum average power at 20 MHz.

   * Andrew LDF4-50A, 1/2" OD, foam-insulated 50 Ω Heliax is rated for 0.300 dB loss per 100 feet and 7.75 kW maximum average power at 20 MHz.
//

Thanks, Rich. This seems to me to indicate that the foam plastic dielectric is carrying heat away from the center conductor. What do you (or does anyone else) think about that theory?

Below is an extract of four pages from the Andrew catalog, including their comments on the subject.

R. Fry

That was really interesting. The higher power rating does seem to be due to better heat transfer away from the center conductor through the foam dielectric.

"Average power ratings for semiflexible cables are determined by the maximum permissible inner conductor temperature. This maximum temperature depends on the type of dielectric... "

Its cheaper than all the RG and LMR stuff since its being sold at or below scrap prices as the 2way biz drys up and towers come down.

Carl

Finally got some time to take another look at this question.  I'm curious about something...


I would like to see the difference in loss between say, 100' of RG-59U at 100 mhz and openwire line using the SAME diameter wire as used in the RG-59/U inner conductor. (.032")


100' of RG-59/U shows about 3.39 db loss at 100 mhz. (1:1 swr)

Using this feedline calculator:
http://vk1od.net/calc/tl/tllc.php


100' of .032" diameter openwire spaced at 1" apart has an impedance of about 412 ohms. However, I didn't find an openwire calculator on the web that shows loss.  Does anyone have a way to calculate the loss for 100' at 100mhz using only air as the dielectric? (and .032" wire)


Actually, one of the the RG-59/U's conductors is fatter because it is shield, so should show a better IR loss.  The openwire uses .032 wires for both leads.

If dielectric has little effect on the loss at 100 mhz, then the loss between the two feedlines should be close, no?  If not, then maybe we are dealing with something else besides dielectric, like distributed capacitance causing circulating "eddy" currents?  The coax obviously has more capacitance between leads per foot than the openwire.

Watsa anyone?

T